[B3D Analysis] R600 has been unleashed upon an unsuspecting enthusiast community

the screenshots of rainbow six vegas were captured on 1.04 patch, i think that is the newest patch that i can downlode from the interneret, so what version patch are you using ?

 

There appears to be a new patch due the 23rd, so I'm guessing ATI's already got a copy.

Just one, though.

 

Our testers used 1.04 and the shadows were fine.

 

Either the drivers or your testers are flawed then. Thats not the only games I've noticed screwed up shadows in.

 

Not true, but it sure seems that on most hardware sites/forums the general assumption is that everyone posting is male. I gave up ranting about being called "he" years ago, since it never seemed to help any.

Back to the topic, personally (and the imagenazis *will* likely get me!) I'd far rather have shadows and HDR than AA. AA is the first thing I kill to get better performance. Don't get me wrong - I like AA, but I like running in a higher resolution and with detail settings more than I like HDR.

Of course, with that said, I haven't had to turn AA off since my upgrade, and also, it is extremely annoying to enable a feature and have the drivers/app/OS ignore it.

 

It`s a bit of a conundrum with AA really...it largely depends upon the game. In some titles, aliasing is hardly bothersome, in others it`s horrid.

 

Maybe I misunderstood but the tested Rainbow 6 with renamed exe to oblivion.exe. Maybe chuck patch had to do something with the shadows and forcing AA.

Zvekan

 

If that's the case, then all bets are off. Forcing applications to take paths that weren't designed or tested for them is completely bogus. If you get incorrect results by doing this, then don't do that! You're essentially telling the driver that the application is Oblivion when it's not. The driver is going to do it's naive thing and try to make AA work. This may trigger bugs because the application is doing effects that Oblivion did not.

So either the R6V shadow bugs were caused by renaming the application (as suggested by Zvekan) or the driver has already been fixed.

 

Ahem...

 

I saw your post, but didn't understand to what it was referring. If you'd mentioned that the R6V had been renamed then it would have clicked

 

See, Dave's typical response using hints and innuendo backfired in this situation! Even when stated clearly, people look for the hidden!

 

He probably should have posted something like, "This just in: RENAMING APPS CAUSES APP-SPECIFIC FIXES TO GO HAYWIRE." Seriously, trying to spin this particular thing into "HD 2900 XT has shoddy drivers" is disingenuous as best.

 

it's been a month now

where's the follow up reviews/articles on IQ and Performance already? every other site has covered IQ and Performance already, some even more than once giving it another chance since new drivers have been released, what's the hold up guys? waiting to see what your further analysis is

*edit*to clarify i think i worded that wrong....i am asking if there will be follow ups to the article since this was a launch review and things are bound to be different a month later

 

Rys has said it will be out shortly. Had his hands full with Computex and everything. Should be out before July hits.

 

Does everyone at work wonder WTF Dave is talking about?

 

OMGhi2u!

...I couldn't help it.

 

Err...should people instead be spinning that the HD 2900XT has spectacular drivers?

Call of Jaurez had noticeable rendering issues. That doesn't automatically translate to "shoddy drivers", and I don't think anyone here tried to "spin" it that way. Also note that the application being renamed in R6:V, *according to the screenshots I saw* seemed only to apply to the 2900XT - there was no notation that such a thing had to be done with the G80 screenshots. Again, that doesn't automatically translate to "shoddy drivers" but if it isn't doing something that G80 is doing...well, I think you can draw the conclusion that they at least need work.

 

Minor caveat:the point is that they renamed the app to get AA working with the 2900 in that particular game through a hack that was aimed at another game, not taking into consideration that R6:V, due to it`s UE3 heritage, isn`t really a prime candidate for having AA on through it`s DX9 path. That alone induces a lot of things that can go wrong that have nothing to do with ATIs driver quality. nV seems to have somehow worked around Vegas's limitation WRT AA somehow, and they offer AA transparently through their drivers(thus no renaming or messing around was necessary, thus limiting the things that could go wrong outside of driver issues with that exact game). That being said, the 2900 has below par drivers ATM, IMHO. And I`m saying that as an user of one. Not to mention that ATIs reluctance when it comes to allowing more flexibility in their user-managed profiles, like nV has, is annoying, especially since I`m ramping up for re-entering the bliss that is Crossfire(which means non scaling games that need renaming, games that artifact due to the multi-chip rendering algorithm employed etc.)

 

That is all I was saying (the arrowed part) - nVidia's implementation didn't have a notation that it needed the application hack in order to render the AA/shadows/etc. while the 2900 XT did. IMHO that is a bigger issue, that you need some sort of a hack to get parts of it to work. Obviously if you are "hacking" apps/drivers/etc. then it isn't supported and you are at your own risk - it is ridiculous to think otherwise. Rendering problems and lack of features of course need to be investigated.

 

AFAIK its the application itself that does not allow AA on R6V - I believe (although I'm not sure) that it uses defered shading which doesn't play with MSAA at all. In this instance not having AA available on the game is a case of folllowing what the game specifies.

(Given the performances seen from NVIDIA's AA, via the control panel, I suspect that they are altering the renderer in oder to provide SuperSampling; but I'm only speculating there)